JPH07105373A - Method and device for classifying fingerprint pattern - Google Patents
Method and device for classifying fingerprint patternInfo
- Publication number
- JPH07105373A JPH07105373A JP5251387A JP25138793A JPH07105373A JP H07105373 A JPH07105373 A JP H07105373A JP 5251387 A JP5251387 A JP 5251387A JP 25138793 A JP25138793 A JP 25138793A JP H07105373 A JPH07105373 A JP H07105373A
- Authority
- JP
- Japan
- Prior art keywords
- lines
- fingerprint
- fingerprint image
- scanning line
- scanning
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Landscapes
- Collating Specific Patterns (AREA)
- Image Input (AREA)
Abstract
Description
【０００１】[0001]
【産業上の利用分野】近年、電算機が社会全般に普及す
るに伴い、安全性（セキュリティ）をいかに確保するか
という点に世間の関心が集まっている。特に、電算機室
への入室や端末利用の際の本人確認の手段として、これ
まで用いられてきたＩＤカードや暗証番号には安全確保
の面から多くの疑問が提起されている。[Industrial application] With the spread of computers throughout society in recent years, public attention has been focused on how to secure safety. In particular, many questions have been raised from the viewpoint of ensuring security for the ID cards and personal identification numbers that have been used so far as means for identifying the person when entering a computer room or using a terminal.
【０００２】一方、指紋は「万人不同」、「終生不変」
という２大特徴を持つため、本人確認の最も有力な手段
として考えられ、指紋を用いた簡便な個人照合システム
に関して多くの研究開発が行われている。本発明は上記
した指紋により個人照合を行う際に必要となる指紋紋様
分類方法および装置に関する。On the other hand, fingerprints are "universal", "lifetime invariant"
Since it has two major characteristics, it is considered to be the most powerful means for personal identification, and much research and development has been conducted on a simple personal identification system using fingerprints. The present invention relates to a fingerprint pattern classification method and device required for individual verification using the above fingerprint.
【０００３】[0003]
【従来の技術】指紋紋様を分類する手法としては、従来
から、指紋の隆線または谷線の線分を算出し、その角度
分布により紋様を分類する方法が提案されている（特開
昭６３−１５３６８７号公報参照）。図１２は上記した
従来の指紋紋様の分類方法を示すフローチャートであ
り、図１３はその処理工程を説明する図、図１４は各紋
様毎の角度分布を示す図である。次に図１２〜図１４に
より従来における指紋紋様の分類方法を説明する。2. Description of the Related Art As a method of classifying fingerprint patterns, a method of calculating line segments of ridges or valleys of fingerprints and classifying patterns by the angle distribution thereof has been proposed (Japanese Patent Laid-Open No. 63-63). -153687). FIG. 12 is a flowchart showing the above-described conventional fingerprint pattern classification method, FIG. 13 is a diagram for explaining the processing steps, and FIG. 14 is a diagram showing the angle distribution for each pattern. Next, a conventional fingerprint pattern classification method will be described with reference to FIGS.
【０００４】図１２のステップＳ１において指紋センサ
により検出された指紋の画像信号から雑音を除去し、ス
テップＳ２において指紋の画像信号を所定のしきい値と
比較して二値化する。ステップＳ３において、二値化さ
れた信号を細線化し、図１３（ａ）に示す指紋の画像デ
ータを得る。ステップＳ４において、細線化された指紋
の画像データより線分列を作成し、ステップＳ５におい
て、作成された線分列の角度を集計する。すなわち、図
１３（ｂ）に示すように細線化された指紋の画像データ
を折れ線で近似することにより線分列を求め、各折れ線
間の角度θを求める。そして、角度θを集計して、図１
３（ｃ）に示すように角度の度数分布を求める。In step S1 of FIG. 12, noise is removed from the fingerprint image signal detected by the fingerprint sensor, and in step S2 the fingerprint image signal is compared with a predetermined threshold value and binarized. In step S3, the binarized signal is thinned to obtain the image data of the fingerprint shown in FIG. In step S4, a line segment sequence is created from the thinned fingerprint image data, and in step S5, the angles of the created line segment sequence are totaled. That is, as shown in FIG. 13B, a line segment sequence is obtained by approximating the image data of the thinned fingerprint with broken lines, and the angle θ between the broken lines is obtained. Then, the angle θ is tabulated and shown in FIG.
As shown in 3 (c), the frequency distribution of angles is obtained.
【０００５】一方、辞書Ｄには、図１４に示すように、
各指紋の紋様の型、すなわち、渦型、左流れ型、右流れ
型、柱型等の指紋紋様の角度の度数分布等が蓄積されて
おり、ステップＳ６において、ステップＳ５において求
めた指紋の紋様の角度集計データと辞書Ｄに蓄積されて
いる各指紋の紋様の度数分布と比較する。そして、ステ
ップＳ５において求めた指紋の紋様を辞書Ｄに格納され
た各指紋の紋様の特徴のうち最も類似した特徴を持つ型
に分類する。On the other hand, in the dictionary D, as shown in FIG.
The fingerprint pattern type of each fingerprint, that is, the frequency distribution of the angles of the fingerprint pattern such as the vortex type, the left flow type, the right flow type, and the column type is accumulated, and the fingerprint pattern obtained in step S5 in step S6. The angle tabulation data and the frequency distribution of the fingerprint pattern stored in the dictionary D are compared. Then, the fingerprint pattern obtained in step S5 is classified into the type having the most similar feature among the fingerprint features stored in the dictionary D.
【０００６】[0006]
【発明が解決しようとする課題】ところで、上記した従
来の方法は、線分列の作成、その角度の集計に時間がか
かり、処理速度が遅いという問題があった。本発明は上
記した従来技術の欠点を改善するためになされたもので
あって、本発明の目的は、より高速にかつ簡便に指紋の
紋様を分類することができる指紋紋様分類方法および装
置を提供することである。By the way, the above-mentioned conventional method has a problem that it takes a long time to create a line segment row and to collect the angles, and the processing speed is slow. The present invention has been made in order to improve the above-mentioned drawbacks of the prior art, and an object of the present invention is to provide a fingerprint pattern classification method and apparatus capable of classifying fingerprint patterns more quickly and easily. It is to be.
【０００７】[0007]
【課題を解決するための手段】図１は本発明の原理図で
ある。同図において、１は指紋像、ｌ1 ，ｒ1 ，ｌ2，
ｒ2 は指紋像の下部（指紋像の指先側を上側とする）を
斜めに走査する走査線、２は指紋を読み取り画像データ
を生成する指紋センサ、３は指紋センサ２により読み取
った指紋の画像データを２値化もしくは細線化する２値
化／細線化手段、４は走査線ｌ1 ，ｒ1 ，ｌ2 ，ｒ2 に
より指紋像１を走査し、走査線と交叉する指紋像の交叉
線数を求める走査手段、５は交叉線数を比較し指紋像の
型を判断する判断手段である。FIG. 1 shows the principle of the present invention. In the figure, 1 is a fingerprint image, l1, r1, l2,
r2 is a scanning line that obliquely scans the lower part of the fingerprint image (the fingertip side of the fingerprint image is the upper side), 2 is a fingerprint sensor that reads the fingerprint and generates image data, and 3 is image data of the fingerprint read by the fingerprint sensor 2. 2 is a binarization / thinning means for binarizing or thinning the scanning line 4. Scanning means 4 scans the fingerprint image 1 with the scanning lines l1, r1, l2, r2 and obtains the number of intersecting lines of the fingerprint image intersecting the scanning lines. Reference numeral 5 is a judging means for judging the type of fingerprint image by comparing the numbers of crossover lines.
【０００８】上記課題を解決するため、本発明の請求項
１の発明は、指紋像１の下部に走査線を設け、上記走査
線と交叉する指紋像の線の数を数えて、その交叉線数の
多寡により指紋の紋様を分類するようにしたものであ
る。本発明の請求項２の発明は、請求項１の発明におい
て、２値化した指紋像を用いて走査線と交叉する線の数
を数えることにより指紋の紋様を分類するようにしたも
のである。In order to solve the above-mentioned problems, the invention of claim 1 of the present invention provides a scanning line below the fingerprint image 1, counts the number of lines of the fingerprint image intersecting the scanning line, and intersects the line. The fingerprint patterns are classified according to the number of numbers. According to a second aspect of the present invention, in the first aspect of the invention, fingerprint patterns are classified by counting the number of lines intersecting the scanning lines using the binarized fingerprint image. .
【０００９】本発明の請求項３の発明は、請求項１の発
明において、細線化した指紋像を用いて走査線と交叉す
る線の数を数えることにより指紋の紋様を分類するよう
にしたものである。本発明の請求項４の発明は、請求項
１の発明において、指紋像１の隆線の交叉線数を数える
ことにより指紋の紋様を分類するようにしたものであ
る。According to a third aspect of the present invention, in the first aspect of the invention, fingerprint patterns are classified by counting the number of lines intersecting the scanning lines using the thinned fingerprint image. Is. According to a fourth aspect of the present invention, in the first aspect of the invention, fingerprint patterns are classified by counting the number of intersecting lines of the ridges of the fingerprint image 1.
【００１０】本発明の請求項５の発明は、請求項１の発
明において、指紋像１の谷線の交叉線数を数えることに
より指紋の紋様を分類するようにしたものである。本発
明の請求項６の発明は、請求項１の発明において、指紋
像１の下方の左の角および右の角に対して三角形を成す
方向にそれぞれ第１および第２の走査線ｌ1 ，ｒ1 を設
けるとともに、指紋像の下方の左の角および右の角もし
くは左右の角の近傍から、指紋像の下辺に対して鋭角を
なす方向にそれぞれ第３および第４の走査線ｌ2 ，ｒ2
を設け、第１および第２の走査線ｌ1 ，ｒ1 と指紋像と
の交叉線数を調べ、第１の走査線ｌ1 との交叉線数が多
く、第２の走査線ｒ1 との交叉線数が少ない場合に、指
紋像の紋様を左流れ型もしくは左袋型に分類し、第１の
走査線ｌ1 との交叉線数が少なく第２の走査線ｒ1 との
交叉線数が多い場合に、指紋像１の紋様を右流れ型もし
くは右袋型に分類し、さらに、第１の走査線ｌ1 との交
叉線数と第２の走査線ｒ1 との交叉線数の差が少ない場
合に、第３および第４の走査線ｌ2 ，ｒ2 と指紋像１と
の交叉線数を調べ、第３の走査線ｌ2 と第４の走査線ｒ
2 と指紋像１との交叉線数が共に多い場合は指紋像１の
紋様を渦型に分類し、第３の走査線ｌ2 と第４の走査線
ｒ2 と指紋像との交叉線数が共に少ない場合は指紋像１
の紋様を弓型に分類するようにしたものである。According to a fifth aspect of the present invention, in the first aspect of the invention, fingerprint patterns are classified by counting the number of intersecting lines of the valley lines of the fingerprint image 1. According to a sixth aspect of the present invention, in the first aspect of the invention, the first and second scanning lines l1 and r1 are formed in directions forming a triangle with respect to the lower left corner and the lower right corner of the fingerprint image 1, respectively. And the third and fourth scanning lines l2 and r2 from the vicinity of the lower left corner and the right corner or the left and right corners of the fingerprint image in the direction forming an acute angle with the lower side of the fingerprint image, respectively.
The number of intersections between the first and second scanning lines l1 and r1 and the fingerprint image is examined, the number of intersections with the first scanning line l1 is large, and the number of intersections with the second scanning line r1 is large. When the number of lines is small, the patterns of the fingerprint image are classified into the left flow type or the left bag type, and when the number of intersecting lines with the first scanning line l1 is small and the number of intersecting lines with the second scanning line r1 is large, If the pattern of the fingerprint image 1 is classified into the right-flow type or the right-bag type, and the difference between the number of crossover lines with the first scanning line l1 and the number of crossover lines with the second scanning line r1 is small, The number of intersecting lines between the third and fourth scanning lines l2 and r2 and the fingerprint image 1 is checked, and the third scanning line l2 and the fourth scanning line r
When the number of crossover lines between 2 and the fingerprint image 1 is large, the pattern of the fingerprint image 1 is classified into a vortex type, and the number of crossover lines between the third scanning line l2, the fourth scanning line r2, and the fingerprint image is both Fingerprint image 1
The patterns are classified as bows.
【００１１】本発明の請求項７の発明は、請求項１の発
明において、指紋像１の下方の左の角および右の角もし
くは左右の角の近傍から、指紋像１の下辺に対して鋭角
をなす方向にそれぞれ第３および第４の走査線ｌ2 ，ｒ
2 を設け、第３および第４の走査線ｌ2 ，ｒ2 と指紋像
１との交叉線数を調べ、第３の走査線ｌ2 と指紋像１と
の交叉線数が少なく、第４の走査線ｒ2 と指紋像１との
交叉線数が多い場合は指紋像１の紋様を左流れ型もしく
は左袋型に分類し、第３の走査線ｌ2 と指紋像１との交
叉線数が多く、第４の走査線ｒ2 と指紋像１との交叉線
数が少ない場合は指紋像１の紋様を右流れ型もしくは右
袋型に分類し、第３の走査線ｌ2 と第４の走査線ｒ2 と
指紋像との交叉線数が共に多い場合は指紋像１の紋様を
渦型に分類し、第３の走査線ｌ2 と第４の走査線ｒ2 と
指紋像１との交叉線数が共に少ない場合は指紋像１の紋
様を弓型に分類するようにしたものである。According to a seventh aspect of the present invention, in the first aspect of the invention, an acute angle is formed from the lower left corner and the right corner or the vicinity of the right and left corners of the fingerprint image 1 to the lower side of the fingerprint image 1. The third and fourth scanning lines l2 and r in the direction
2 is provided to check the number of crossover lines between the third and fourth scanning lines l2 and r2 and the fingerprint image 1, and the number of crossover lines between the third scanning line l2 and the fingerprint image 1 is small and the fourth scanning line When the number of crossover lines between r2 and the fingerprint image 1 is large, the pattern of the fingerprint image 1 is classified into the left flow type or the left bag type, and the number of crossover lines between the third scanning line l2 and the fingerprint image 1 is large, When the number of crossover lines between the scanning line r2 of 4 and the fingerprint image 1 is small, the pattern of the fingerprint image 1 is classified into the right flow type or the right bag type, and the third scanning line l2, the fourth scanning line r2, and the fingerprint. If the number of lines of intersection with the image is large, the pattern of the fingerprint image 1 is classified into a vortex type, and if the number of lines of intersection of the third scanning line l2, the fourth scanning line r2 and the fingerprint image 1 is both small. The patterns of the fingerprint image 1 are classified into bow shapes.
【００１２】本発明の請求項８の発明は、請求項７また
は請求項８の発明において、一方の走査線と指紋像１と
の交叉線数と他方の走査線と指紋像１との交叉線数との
差を求め、求めた差が閾値以上か、あるいは差が大であ
るとき、一方の走査線と指紋像１との交叉線数が多く、
他方の走査線と指紋像１との交叉線数が少ないと判断す
るようにしたものである。The invention of claim 8 of the present invention is the invention of claim 7 or 8, wherein the number of intersections between one scanning line and fingerprint image 1 and the intersections between the other scanning line and fingerprint image 1 When the difference between the number of scan lines and the fingerprint image 1 is larger than the threshold value or the difference is large, the number of cross lines between one scanning line and the fingerprint image 1 is large,
It is determined that the number of intersecting lines between the other scanning line and the fingerprint image 1 is small.
【００１３】本発明の請求項９の発明は、請求項７また
は請求項８の発明において、一方の走査線と指紋像１と
の交叉線数と他方の走査線と指紋像１との交叉線数との
比を求め、求めた比が閾値以上か、あるいは比が大であ
るとき、一方の走査線と指紋像１との交叉線数が多く、
他方の走査線と指紋像１との交叉線数が少ないと判断す
るようにしたものである。According to a ninth aspect of the present invention, in the invention according to the seventh or eighth aspect, the number of intersecting lines between one scanning line and fingerprint image 1 and the intersecting line between the other scanning line and fingerprint image 1 If the calculated ratio is greater than or equal to the threshold value or the ratio is large, the number of lines intersecting one scanning line with the fingerprint image 1 is large,
It is determined that the number of intersecting lines between the other scanning line and the fingerprint image 1 is small.
【００１４】本発明の請求項１０の発明は、指紋像を読
み取る指紋センサ２と、指紋センサ２により読み取った
指紋像１を２値化もしくは細線化する２値化／細線化手
段３と、２値化もしくは細線化された指紋像１の下部を
走査し、走査線ｌ1 ，ｒ1 ，ｌ2 ，ｒ2 と指紋像１との
交叉線数を求める走査手段４と、求めた交叉線数を各走
査線毎に比較し、交叉線数の多寡により指紋の紋様を判
断する判断手段５とから指紋紋様分類装置を構成したも
のである。According to a tenth aspect of the present invention, a fingerprint sensor 2 for reading a fingerprint image, a binarizing / thinning means 3 for binarizing or thinning the fingerprint image 1 read by the fingerprint sensor 2, and 2 The scanning unit 4 for scanning the lower part of the digitized or thinned fingerprint image 1 to determine the number of intersections of the scanning lines l1, r1, l2, r2 and the fingerprint image 1, and the determined number of intersections for each scanning line. The fingerprint pattern classifying device is configured by a judgment means 5 for judging the fingerprint pattern based on the number of crossover lines.
【００１５】[0015]
【作用】図２（ａ）は本発明における走査線ｌ1 ，ｒ1
を示し，（ｂ）は本発明における走査線ｌ2 ，ｒ2 を示
している。本発明においては、図２（ａ），（ｂ）に示
す走査線ｌ1 ，ｒ1 ，ｌ2 ，ｒ2 と交叉する指紋の隆線
（谷線）を数え、その交叉する数に応じて指紋の紋様を
分類する。2A shows the scanning lines l1 and r1 in the present invention.
And (b) shows scanning lines l2 and r2 in the present invention. In the present invention, the ridges (valley lines) of the fingerprints intersecting the scanning lines l1, r1, l2, r2 shown in FIGS. 2 (a) and 2 (b) are counted, and the fingerprint pattern is formed according to the number of intersections. Classify.
【００１６】交叉する指紋の隆線（谷線）を数える方法
としては、指紋の画像データを２値化した場合には隆線
の画素を１、谷線の画素を０とし、また、指紋の画像デ
ータを細線化した場合には、細線を１、その他を０とし
て、走査線の隣合う画素間の差が、１または−１になる
箇所の数を数える方法を用いることができる。図２
（ａ）に示す走査線ｌ1 ，ｒ1 を用いた場合、指紋の紋
様が、例えば左流れ型では、図３（ａ）に示すように、
左側の交叉線数が多く、右側の交叉線数が少なくなる。
また、指紋の紋様が渦型では、図３（ｂ）に示すように
左右の交叉線数がほぼ同等となる。すなわち、指紋の紋
様の型に応じて、左右の走査線ｌ1，ｒ1 との交叉線数
は図４に示すようになる。As a method of counting the ridges (troughs) of the intersecting fingerprints, when the image data of the fingerprint is binarized, the pixel of the ridge is set to 1, the pixel of the valley is set to 0, and the fingerprint When the image data is thinned, a method can be used in which the thin line is set to 1 and the others are set to 0, and the number of places where the difference between adjacent pixels of the scanning line becomes 1 or -1. Figure 2
When the scanning lines l1 and r1 shown in (a) are used, if the fingerprint pattern is, for example, a left-flow pattern, as shown in FIG.
The number of crossover lines on the left side is large, and the number of crossover lines on the right side is small.
Further, when the fingerprint pattern is a spiral pattern, the number of crossover lines on the left and right becomes substantially the same as shown in FIG. That is, the number of intersecting lines with the left and right scanning lines l1 and r1 is as shown in FIG. 4 according to the pattern of the fingerprint.
【００１７】また、図２（ｂ）に示す走査線ｌ2 ，ｒ2
を用いた場合、指紋の紋様が、渦型では、図３（ｃ）に
示すように、左右の交叉線数が共に多く、また、弓型で
は、図３（ｄ）に示すように左右の交叉線数が共に少な
くなる。すなわち、指紋の紋様の型に応じて、左右の走
査線との交叉線数は図５に示すようになる。したがっ
て、上記のように図２（ａ），（ｂ）に示す走査線ｌ1
，ｒ1 ，ｌ2，ｒ2 を用いて指紋の紋様との交叉線数を
数え、交叉線数に対して適当な閾値を設けることによ
り、図７に示すような紋様の分類が可能となる。Further, scanning lines l2 and r2 shown in FIG.
When the fingerprint pattern is used, when the fingerprint pattern is a vortex type, the left and right crossover lines are both large as shown in FIG. Both the number of crossover lines is reduced. That is, the number of intersecting lines with the left and right scanning lines is as shown in FIG. 5 according to the pattern of the fingerprint. Therefore, as described above, the scanning line l1 shown in FIGS.
, R1, l2, r2 are used to count the number of intersecting lines with the fingerprint pattern and an appropriate threshold value is set for the number of intersecting lines, whereby the pattern can be classified as shown in FIG.
【００１８】さらに、図２（ｂ）に示す走査線ｌ2 ，ｒ
2 のみを用いた場合には、図６に示すように走査線ｌ2
，ｒ2 と指紋の隆線（谷線）が交叉し、左右の走査線
との交叉線数は図８に示すようになることも予想され
る。したがって、図２（ｂ）に示す走査線ｌ2 ，ｒ2 の
みを用いた場合には、図９に示すように指紋の紋様を分
類することも可能である。Further, scanning lines l2 and r shown in FIG.
When only 2 is used, as shown in FIG.
, R2 and the ridges (valleys) of the fingerprint intersect, and it is expected that the number of intersections with the left and right scanning lines will be as shown in FIG. Therefore, when only the scanning lines l2 and r2 shown in FIG. 2B are used, it is also possible to classify fingerprint patterns as shown in FIG.
【００１９】本発明の請求項１〜１０の発明においては
上記原理に基づき、指紋の紋様を左流れ型、右流れ型、
渦型、弓型に分類するようにしたので、従来技術のよう
の、線分列の作成、角度集計等の処理を行う必要がな
く、高速かつ簡便に指紋の紋様を分類することができ
る。According to the first to tenth aspects of the present invention, based on the above principle, the fingerprint pattern is left-flow type, right-flow type,
Since the patterns are classified into the vortex type and the bow type, it is possible to classify fingerprint patterns at high speed and easily without the need to perform processing such as line segment sequence and angle aggregation unlike the prior art.
【００２０】[0020]
【実施例】図１０は本発明の実施例を示すブロック図で
ある。同図において、１１は指紋の画像を読み取り電気
信号に変換する指紋センサ、１２は読み取った指紋の濃
淡画像データを記憶する濃淡画像記憶手段、１３は濃淡
画像記憶手段１２に記憶された指紋の濃淡画像データを
所定の閾値と比較して１または０のデジタルデータに２
値化する２値化回路、１４は２値化された画像データを
記憶する２値化像記憶手段、１５は２値化された画像デ
ータを細線化する細線化回路、１６は細線化された画像
データを記憶する細線化像記憶手段である。FIG. 10 is a block diagram showing an embodiment of the present invention. In the figure, 11 is a fingerprint sensor for reading an image of a fingerprint and converting it into an electric signal, 12 is a grayscale image storage means for storing grayscale image data of the read fingerprint, and 13 is a grayscale of the fingerprint stored in the grayscale image storage means 12. Image data is compared with a predetermined threshold value and digital data of 1 or 0 is converted into 2
A binarizing circuit for binarizing, 14 is a binarized image storing means for storing binarized image data, 15 is a thinning circuit for thinning the binarized image data, and 16 is a thin line. It is a thinned image storage means for storing image data.
【００２１】また、１７は前記した図２（ａ），（ｂ）
に示す走査線ｌ1 ，ｒ1 ，ｌ2 ，ｒ2 により指紋の紋様
の画像データを走査し交叉線数を求める走査回路であ
り、本発明は、指紋の２値化像、細線化像のいずれの画
像データにも適用できるので、図１０の実施例において
は、走査回路１７が２値化像記憶手段１４と細線化像記
憶手段１６の両方と結線した例が示されている。Further, 17 is the above-mentioned FIG. 2 (a), (b)
The scanning circuit for scanning the image data of the fingerprint pattern by scanning lines l1, r1, l2, and r2 shown in FIG. 2 to find the number of crossover lines. The present invention is applicable to either binary image data or thin line image data of a fingerprint. Since it can also be applied to, the embodiment of FIG. 10 shows an example in which the scanning circuit 17 is connected to both the binarized image storage means 14 and the thinned image storage means 16.
【００２２】１８は走査回路１７により求めた交叉線数
を各走査線について比較する交叉線数比較回路、１９は
交叉線数比較回路１８における比較結果に基づき指紋の
紋様の型を判断する判断回路である。図１１は本実施例
における処理を示すフローチャートであり、同図を参照
して図１０に示した実施例の動作を説明する。Reference numeral 18 is a crossing line number comparing circuit for comparing the number of crossing lines obtained by the scanning circuit 17 with respect to each scanning line, and 19 is a judging circuit for judging the pattern type of the fingerprint based on the comparison result in the crossing line number comparing circuit 18. Is. FIG. 11 is a flow chart showing the processing in this embodiment, and the operation of the embodiment shown in FIG. 10 will be described with reference to this figure.
【００２３】指紋センサ１１により指紋の画像を読み取
り濃淡画像として濃淡画像記憶手段１２に記憶し（図１
１のステップＳ１）、２値化回路１３において１，０の
デジタルデータに２値化し、２値化像記憶手段１４に記
憶する（図１１のステップＳ２）。ついで、２値化され
た画像データを細線化回路１５において細線化し細線化
像記憶手段１６に記憶する（図１１のステップＳ３）。The fingerprint image is read by the fingerprint sensor 11 and stored in the grayscale image storage means 12 as a grayscale image (see FIG. 1).
In step S1 of 1), the binarization circuit 13 binarizes the digital data of 1,0 and stores it in the binarized image storage means 14 (step S2 of FIG. 11). Then, the binarized image data is thinned by the thinning circuit 15 and stored in the thinned image storage means 16 (step S3 in FIG. 11).
【００２４】走査回路１７は２値化像記憶手段１４もし
くは細線化像記憶手段１６に記憶された画像データのい
ずれかの画像データを図２（ａ），（ｂ）に示した走査
線ｌ1 ，ｒ1 ，ｌ2 ，ｒ2 で走査し、各走査線について
の交叉線数を求める（図１１のステップＳ４）。交叉線
数比較回路１８は走査回路１７において求めた交叉線数
を比較し（図１１のステップＳ５）、判断回路１９が前
記した図４，５もしくは図８に示した手法により指紋の
紋様の型を判断する（図１１のステップＳ６）。The scanning circuit 17 scans the image data of either the image data stored in the binarized image storage means 14 or the thinned image storage means 16 with the scanning line l1 shown in FIGS. 2 (a) and 2 (b). Scanning is performed with r1, l2, and r2 to obtain the number of intersecting lines for each scanning line (step S4 in FIG. 11). The crossover line number comparing circuit 18 compares the crossover line numbers obtained in the scanning circuit 17 (step S5 in FIG. 11), and the judging circuit 19 uses the method shown in FIG. Is determined (step S6 in FIG. 11).
【００２５】上記交叉線数比較回路１８および判断回路
１９において指紋の紋様の型の判断は次のいずれかの手
法により行うことができる。下記の実施例において、図
２（ａ）の左側の走査線ｌ1 との交叉線数をＬ1 、右側
の走査線ｒ1 との交叉線数をＲ1 とする。また、図２
（ｂ）の左側の走査線ｌ2 との交叉線数をＬ2 とし、右
側の走査線ｒ2 との交叉線数をＲ2 とし、これらの交叉
線数もしくはこれらの差、商等と所定の閾値TH1 ，TH2
，TH3 ，TH4 ，TH5 ，TH6 の大小関係により次のよう
に指紋の紋様の型の判断を行う。 （１）実施例１ 本実施例は図２（ａ），（ｂ）に示した走査線ｌ1 ，ｒ
1 ，ｌ2 ，ｒ2 を用い、図４，５および図７に示した手
法により指紋の紋様を分類する。 左流れの判定： Ｌ1 ＞TH1 ，Ｒ1 ＜TH2 右流れの判定： Ｌ1 ＜TH2 ，Ｒ1 ＞TH1 渦・弓型の判定：｜Ｌ1 −Ｒ1 ｜＜TH3 渦型の判定 ：Ｌ2 ＞TH1 ，Ｒ2 ＞TH1 弓型の判定 ：Ｌ2 ＜TH2 ，Ｒ2 ＜TH2 （２）実施例２ 本実施例は図２（ａ），（ｂ）に示した走査線ｌ1 ，ｒ
1 ，ｌ2 ，ｒ2 を用い、図４，５および図７に示した手
法により指紋の紋様を分類する。 左流れの判定： Ｌ1 −Ｒ1 ＞TH4 右流れの判定： Ｒ1 −Ｌ1 ＞TH4 渦・弓型の判定：｜Ｌ1 −Ｒ1 ｜＜TH3 渦型の判定 ：Ｌ2 ＞TH1 ，Ｒ2 ＞TH1 弓型の判定 ：Ｌ2 ＜TH2 ，Ｒ2 ＜TH2 （３）実施例３ 本実施例は図２（ａ），（ｂ）に示した走査線ｌ1 ，ｒ
1 ，ｌ2 ，ｒ2 を用い、図４，５および図７に示した手
法により指紋の紋様を分類する。 左流れの判定： Ｌ1 ／Ｒ1 ＞TH5 右流れの判定： Ｒ1 ／Ｌ1 ＞TH5 渦・弓型の判定：TH6 ＜Ｌ1 ／Ｒ1 ＜TH5 渦型の判定 ：Ｌ2 ＞TH1 ，Ｒ2 ＞TH1 弓型の判定 ：Ｌ2 ＜TH2 ，Ｒ2 ＜TH2 （４）実施例４ 本実施例は図２（ｂ）に示した走査線ｌ2 ，ｒ2 のみを
用い、図８および図９に示した手法により指紋の紋様を
分類する。 左流れの判定： Ｌ2 ＜TH2 ，Ｒ2 ＞TH1 右流れの判定： Ｌ2 ＞TH1 ，Ｒ2 ＜TH2 渦型の判定 ：Ｌ2 ＞TH1 ，Ｒ2 ＞TH1 弓型の判定 ：Ｌ2 ＜TH2 ，Ｒ2 ＜TH2 （５）実施例５ 本実施例は図２（ｂ）に示した走査線ｌ2 ，ｒ2 のみを
用い、図８および図９に示した手法により指紋の紋様を
分類する。 左流れの判定： Ｒ2 −Ｌ2 ＞TH4 右流れの判定： Ｌ2 −Ｒ2 ＞TH4 渦型の判定 ：Ｌ2 ＞TH1 ，Ｒ2 ＞TH1 弓型の判定 ：Ｌ2 ＜TH2 ，Ｒ2 ＜TH2 （６）実施例６ 本実施例は図２（ｂ）に示した走査線ｌ2 ，ｒ2 のみを
用い、図８および図９に示した手法により指紋の紋様を
分類する。 左流れの判定： Ｒ2 ／Ｌ2 ＞TH5 右流れの判定： Ｌ2 ／Ｒ2 ＞TH5 渦型の判定 ：Ｌ2 ＞TH1 ，Ｒ2 ＞TH1 弓型の判定 ：Ｌ2 ＜TH2 ，Ｒ2 ＜TH2The type of fingerprint pattern can be determined by the crossover line number comparing circuit 18 and the determining circuit 19 by one of the following methods. In the following embodiment, the number of intersections with the scanning line l1 on the left side of FIG. 2A is L1, and the number of intersections with the scanning line r1 on the right side is R1. Also, FIG.
The number of intersections with the scanning line l2 on the left side of (b) is L2, the number of intersections with the scanning line r2 on the right side is R2, and the number of these intersections or their difference, quotient and a predetermined threshold value TH1, TH2
, TH3, TH4, TH5, TH6 are compared to determine the fingerprint pattern type as follows. (1) Example 1 In this example, the scanning lines l1 and r shown in FIGS.
The fingerprint patterns are classified by the method shown in FIGS. 4, 5 and 7 using 1, l2 and r2. Judgment of left flow: L1> TH1, R1 <TH2 Judgment of right flow: L1 <TH2, R1> TH1 Vortex / bow type judgment: | L1 -R1 | <TH3 Vortex type judgment: L2> TH1, R2> TH1 Bow type determination: L2 <TH2, R2 <TH2 (2) Example 2 In this example, the scanning lines l1 and r shown in FIGS. 2 (a) and 2 (b) are used.
The fingerprint patterns are classified by the method shown in FIGS. 4, 5 and 7 using 1, l2 and r2. Judgment of left flow: L1 -R1> TH4 Judgment of right flow: R1 -L1> TH4 Judgment of vortex / bow: | L1 -R1 | <TH3 Judgment of vortex: L2> TH1, R2> TH1 Judgment of bow : L2 <TH2, R2 <TH2 (3) Third Embodiment In this embodiment, the scanning lines l1 and r shown in FIGS. 2 (a) and 2 (b) are used.
The fingerprint patterns are classified by the method shown in FIGS. 4, 5 and 7 using 1, l2 and r2. Left flow judgment: L1 / R1> TH5 Right flow judgment: R1 / L1> TH5 Vortex / bow type judgment: TH6 <L1 / R1 <TH5 Vortex type judgment: L2> TH1, R2> TH1 Bow type judgment : L2 <TH2, R2 <TH2 (4) Fourth Embodiment In this embodiment, fingerprint patterns are classified by the method shown in FIGS. 8 and 9 using only the scanning lines l2 and r2 shown in FIG. 2B. To do. Left flow judgment: L2 <TH2, R2> TH1 Right flow judgment: L2> TH1, R2 <TH2 Vortex type judgment: L2> TH1, R2> TH1 Bow type judgment: L2 <TH2, R2 <TH2 (5 Example 5 In this example, fingerprint patterns are classified by the method shown in FIGS. 8 and 9 using only the scanning lines l2 and r2 shown in FIG. 2B. Judgment of left flow: R2-L2> TH4 Judgment of right flow: L2-R2> TH4 Judgment of vortex type: L2> TH1, R2> TH1 Judgment of bow type: L2 <TH2, R2 <TH2 (6) Example 6 In the present embodiment, only the scanning lines l2 and r2 shown in FIG. 2B are used, and the fingerprint patterns are classified by the method shown in FIGS. Left flow judgment: R2 / L2> TH5 Right flow judgment: L2 / R2> TH5 Vortex type judgment: L2> TH1, R2> TH1 Bow type judgment: L2 <TH2, R2 <TH2
【００２６】[0026]
【発明の効果】以上説明したように、指紋像の下部に走
査線を設け、上記走査線と交叉する指紋像の線の数を数
えて、その交叉線数の多寡により指紋の紋様を分類する
ようにしたので、高速かつ簡便に指紋の紋様を分類する
ことができる。As described above, the scanning line is provided below the fingerprint image, the number of lines of the fingerprint image intersecting with the scanning line is counted, and the fingerprint pattern is classified according to the number of the intersecting lines. Since this is done, the fingerprint patterns can be classified quickly and easily.
【図１】本発明の原理図である。FIG. 1 is a principle diagram of the present invention.
【図２】本発明における走査線を示す図である。FIG. 2 is a diagram showing scanning lines in the present invention.
【図３】指紋の紋様と交叉線数との関係を示す図であ
る。FIG. 3 is a diagram showing a relationship between a fingerprint pattern and the number of crossover lines.
【図４】図２（ａ）の走査線による交叉線数の比較結果
を示す図である。FIG. 4 is a diagram showing a comparison result of the number of crossover lines by the scanning lines of FIG.
【図５】図２（ｂ）の走査線を用いた場合の交叉線数の
比較結果を示す図である。FIG. 5 is a diagram showing a comparison result of the number of crossover lines when the scanning line in FIG. 2B is used.
【図６】図２（ｂ）の走査線における紋様と交叉線数と
の関係を示す図である。FIG. 6 is a diagram showing the relationship between the pattern and the number of crossover lines in the scanning line of FIG. 2 (b).
【図７】指紋の紋様を分類する流れ図である。FIG. 7 is a flowchart for classifying fingerprint patterns.
【図８】図２（ｂ）の走査線のみによる交叉線数の比較
結果を示す図である。FIG. 8 is a diagram showing a comparison result of the number of crossover lines only by the scanning lines in FIG.
【図９】図２（ｂ）の走査線のみにより指紋の紋様を分
類する流れ図である。FIG. 9 is a flowchart for classifying fingerprint patterns by using only the scanning lines shown in FIG. 2B.
【図１０】本発明の実施例のブロック図である。FIG. 10 is a block diagram of an embodiment of the present invention.
【図１１】本発明の実施例のフローチャートである。FIG. 11 is a flowchart of an example of the present invention.
【図１２】従来例のフローチャートである。FIG. 12 is a flowchart of a conventional example.
【図１３】従来例における処理工程を説明する図であ
る。FIG. 13 is a diagram illustrating processing steps in a conventional example.
【図１４】指紋の各紋様毎の角度分布を示す図である。FIG. 14 is a diagram showing an angular distribution of fingerprints for each pattern.
１ 指紋像 ｌ1 ，ｒ1 ，ｌ2 ，ｒ2 走査線 ２，１１ 指紋センサ ３ ２値化／細線化手段 ４ 走査手段 ５ 判断手段 １２ 濃淡画像記憶手段 １３ ２値化回路 １４ ２値化像記憶手段 １５ 細線化回路 １６ 細線化像記憶手段 １７ 走査回路 １８ 交叉線数比較回路 １９ 判断回路 1 Fingerprint image l1, r1, l2, r2 Scanning line 2,11 Fingerprint sensor 3 Binarizing / thinning means 4 Scanning means 5 Judging means 12 Grayscale image storing means 13 Binarizing circuit 14 Binarized image storing means 15 Thin line Circuit 16 thinned image storage means 17 scanning circuit 18 crossover line number comparison circuit 19 judgment circuit
Claims (10)
叉線数の多寡により指紋の紋様を分類することを特徴と
する指紋紋様分類方法。1. A scanning line is provided below the fingerprint image (1), the number of lines of the fingerprint image intersecting with the scanning line is counted, and fingerprint patterns are classified according to the number of intersecting lines. Fingerprint pattern classification method.
する線の数を数えることにより指紋の紋様を分類するこ
とを特徴とする請求項１の指紋紋様分類方法。2. The fingerprint pattern classification method according to claim 1, wherein the fingerprint patterns are classified by counting the number of lines intersecting the scanning lines using the binarized fingerprint image.
する線の数を数えることにより指紋の紋様を分類するこ
とを特徴とする請求項１の指紋紋様分類方法。3. The fingerprint pattern classification method according to claim 1, wherein the fingerprint patterns are classified by counting the number of lines intersecting the scanning lines using the thinned fingerprint image.
とにより指紋の紋様を分類することを特徴とする請求項
１の指紋紋様分類方法。4. The fingerprint pattern classification method according to claim 1, wherein the fingerprint patterns are classified by counting the number of intersecting lines of the ridges of the fingerprint image (1).
とにより指紋の紋様を分類することを特徴とする請求項
１の指紋紋様分類方法。5. The fingerprint pattern classification method according to claim 1, wherein the fingerprint patterns are classified by counting the number of intersecting lines of the valley lines of the fingerprint image (1).
に対して三角形を成す方向にそれぞれ第１および第２の
走査線(l1,r1) を設けるとともに、 指紋像の下方の左の角および右の角もしくは左右の角の
近傍から、指紋像の下辺に対して鋭角をなす方向にそれ
ぞれ第３および第４の走査線(l1,r1) を設け、 第１および第２の走査線(l1,r1) と指紋像との交叉線数
を調べ、第１の走査線(l1)との交叉線数が多く、第２の
走査線(r1)との交叉線数が少ない場合に、指紋像の紋様
を左流れ型もしくは左袋型に分類し、 第１の走査線(l1)との交叉線数が少なく第２の走査線(r
1)との交叉線数が多い場合に、指紋像(1) の紋様を右流
れ型もしくは右袋型に分類し、 さらに、第１の走査線(l1)との交叉線数と第２の走査線
(r1)との交叉線数の差が少ない場合に、第３および第４
の走査線(l2,r2) と指紋像(1) との交叉線数を調べ、第
３の走査線(l2)と第４の走査線(r2)と指紋像(1) との交
叉線数が共に多い場合は指紋像(1) の紋様を渦型に分類
し、 第３の走査線(l2)と第４の走査線(r2)と指紋像との交叉
線数が共に少ない場合は指紋像(1) の紋様を弓型に分類
することを特徴とする請求項１の指紋紋様分類方法。6. The first and second scanning lines (l1, r1) are respectively provided in the directions forming a triangle with respect to the lower left corner and the lower right corner of the fingerprint image (1), and the lower part of the fingerprint image (1) is provided. The third and fourth scanning lines (l1, r1) are provided from the vicinity of the left corner and the right corner or the left and right corners of the fingerprint image in the direction forming an acute angle with the lower side of the fingerprint image, and the first and second scanning lines Check the number of lines intersecting the scanning line (l1, r1) with the fingerprint image, the number of lines intersecting with the first scanning line (l1) is large, and the number of lines intersecting with the second scanning line (r1) is small. In this case, the fingerprint image pattern is classified into the left flow type or the left bag type, and the number of lines intersecting with the first scanning line (l1) is small and the second scanning line (r1
When the number of lines intersecting with 1) is large, the pattern of the fingerprint image (1) is classified into the right flow type or the right bag type, and the number of lines intersecting with the first scanning line (l1) and the second Scan line
If the difference in the number of crossover lines from (r1) is small, the 3rd and 4th
Check the number of crossover lines between the scan line (l2, r2) and the fingerprint image (1), and check the number of crossover lines between the third scan line (l2), the fourth scan line (r2) and the fingerprint image (1). If the number of both is large, the pattern of the fingerprint image (1) is classified into a vortex type, and if the number of intersecting lines of the third scanning line (l2), the fourth scanning line (r2) and the fingerprint image is small, the fingerprint is The fingerprint pattern classification method according to claim 1, wherein the patterns of the image (1) are classified into an arch pattern.
もしくは左右の角の近傍から、指紋像(1) の下辺に対し
て鋭角をなす方向にそれぞれ第３および第４の走査線(l
2,r2) を設け、 第３および第４の走査線(l2,r2) と指紋像(1) との交叉
線数を調べ、第３の走査線(l2)と指紋像(1) との交叉線
数が少なく、第４の走査線(r2)と指紋像(1) との交叉線
数が多い場合は指紋像(1) の紋様を左流れ型もしくは左
袋型に分類し、 第３の走査線(l2)と指紋像(1) との交叉線数が多く、第
４の走査線(r2)と指紋像(1) との交叉線数が少ない場合
は指紋像(1) の紋様を右流れ型もしくは右袋型に分類
し、 第３の走査線(l2)と第４の走査線(r2)と指紋像との交叉
線数が共に多い場合は指紋像(1) の紋様を渦型に分類
し、 第３の走査線(l2)と第４の走査線(r2)と指紋像(1) との
交叉線数が共に少ない場合は指紋像(1) の紋様を弓型に
分類することを特徴とする請求項１の指紋紋様分類方
法。7. The third and fourth corners of the fingerprint image (1) from the vicinity of the lower left corner and the right corner or the left and right corners of the fingerprint image (1) to the lower side of the fingerprint image (1), respectively. Scan line (l
2, r2) is provided, and the number of crossover lines between the third and fourth scanning lines (l2, r2) and the fingerprint image (1) is checked, and the third scanning line (l2) and the fingerprint image (1) When the number of crossover lines is small and the number of crossover lines between the fourth scanning line (r2) and the fingerprint image (1) is large, the pattern of the fingerprint image (1) is classified into the left flow type or the left bag type, and the third If the number of crossover lines between the scanning line (l2) and the fingerprint image (1) is large and the number of crossover lines between the fourth scanning line (r2) and the fingerprint image (1) is small, the pattern of the fingerprint image (1) Are classified as right-flow type or right-bag type, and if the number of crossover lines between the third scanning line (l2), the fourth scanning line (r2), and the fingerprint image is large, the pattern of the fingerprint image (1) is The pattern of the fingerprint image (1) is arched when the number of intersections of the third scanning line (l2), the fourth scanning line (r2) and the fingerprint image (1) is small. The fingerprint pattern classification method according to claim 1, wherein the classification is performed.
と他方の走査線と指紋像(1) との交叉線数との差を求
め、求めた差が閾値以上か、あるいは差が大であると
き、一方の走査線と指紋像(1) との交叉線数が多く、他
方の走査線と指紋像(1) との交叉線数が少ないと判断す
ることを特徴とする請求項６または請求項７の指紋紋様
分類方法。8. A difference between the number of crossover lines between one scanning line and the fingerprint image (1) and the number of crossover lines between the other scanning line and the fingerprint image (1) is calculated, and whether the obtained difference is a threshold value or more, Alternatively, when the difference is large, one of the scanning lines and the fingerprint image (1) has a large number of intersecting lines, and the other scanning line and the fingerprint image (1) have a small number of intersecting lines. The fingerprint pattern classification method according to claim 6 or 7.
と他方の走査線と指紋像(1) との交叉線数との比を求
め、求めた比が閾値以上か、あるいは比が大であると
き、一方の走査線と指紋像(1) との交叉線数が多く、他
方の走査線と指紋像(1) との交叉線数が少ないと判断す
ることを特徴とする請求項６または請求項７の指紋紋様
分類方法。9. A ratio between the number of intersecting lines between one scanning line and the fingerprint image (1) and the number of intersecting lines between the other scanning line and the fingerprint image (1), and whether the obtained ratio is equal to or more than a threshold value, Alternatively, when the ratio is large, it is determined that the number of cross lines between one scanning line and the fingerprint image (1) is large, and the number of cross lines between the other scanning line and the fingerprint image (1) is small. The fingerprint pattern classification method according to claim 6 or 7.
しくは細線化する２値化／細線化手段(3) と、 ２値化もしくは細線化された指紋像(1) の下部を走査
し、走査線(l1,r1,l2,r2) と指紋像(1) との交叉線数を
求める走査手段(4) と、 求めた交叉線数を各走査線毎に比較し、交叉線数の多寡
により指紋の紋様を判断する判断手段(5) とを備えたこ
とを特徴とする指紋紋様分類装置。10. A fingerprint sensor (2) for reading a fingerprint image, and a binarizing / thinning means (3) for binarizing or thinning the fingerprint image (1) read by the fingerprint sensor (2); Scan the lower part of the digitized or thinned fingerprint image (1) and find the number of lines of intersection between the scanning lines (l1, r1, l2, r2) and the fingerprint image (1). A fingerprint pattern classifying device, comprising: a determination means (5) for comparing the number of intersecting lines for each scanning line and determining the fingerprint pattern based on the number of intersecting lines.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP5251387A JPH07105373A (en) | 1993-10-07 | 1993-10-07 | Method and device for classifying fingerprint pattern |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP5251387A JPH07105373A (en) | 1993-10-07 | 1993-10-07 | Method and device for classifying fingerprint pattern |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| JPH07105373A true JPH07105373A (en) | 1995-04-21 |
Family
ID=17222089
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP5251387A Withdrawn JPH07105373A (en) | 1993-10-07 | 1993-10-07 | Method and device for classifying fingerprint pattern |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JPH07105373A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7133542B2 (en) | 2000-09-29 | 2006-11-07 | Chuo Hatsujo Kabushiki Kaisha | Fingerprint verification device and fingerprint verification method |
-
1993
- 1993-10-07 JP JP5251387A patent/JPH07105373A/en not_active Withdrawn
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7133542B2 (en) | 2000-09-29 | 2006-11-07 | Chuo Hatsujo Kabushiki Kaisha | Fingerprint verification device and fingerprint verification method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JPH0823890B2 (en) | Method and apparatus for automatically verifying signature | |
| US7035444B2 (en) | System for fingerprint authentication based of ridge shape | |
| CA2091997C (en) | Character recognition methods including locating and extracting predetermined and apparatus data from a document | |
| JPH02306386A (en) | Character recognizing device | |
| JP4245093B2 (en) | Fingerprint verification device and processing method thereof | |
| JPH07105373A (en) | Method and device for classifying fingerprint pattern | |
| JP3505713B2 (en) | Curve identification system | |
| JP2797848B2 (en) | Optical character reader | |
| WO1992006447A2 (en) | Character recognition methods including separating and extractingparticular portions from extracted data | |
| JP2790689B2 (en) | Fingerprint center position calculation method | |
| JPH05189546A (en) | Device for discriminating authenticity of fingerprint featured point | |
| JP2600680B2 (en) | Personal verification device | |
| JP2868909B2 (en) | Fingerprint collation device | |
| JPH05197792A (en) | Binarizing device for finger print collating device | |
| JP2611258B2 (en) | Dictionary compression method of partial image matching method with variable relative position | |
| JP2871161B2 (en) | Pseudo feature point identification method | |
| JPH03269780A (en) | Fingerprint data binarizing method | |
| JPH06139338A (en) | Fingerprint pattern classifying device | |
| JP4223295B2 (en) | License plate automatic recognition device | |
| JPH05242224A (en) | Fingerprint collating device | |
| JP2894111B2 (en) | Comprehensive judgment method of recognition result in optical type character recognition device | |
| JPH05205035A (en) | Fingerprint collating device | |
| JP3033595B2 (en) | Fingerprint image registration method | |
| JPH05197791A (en) | Finger print collating device | |
| JP2871233B2 (en) | Binary image correction method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| A300 | Application deemed to be withdrawn because no request for examination was validly filed |
Free format text: JAPANESE INTERMEDIATE CODE: A300 Effective date: 20001226 |